Refer to the scenario described in Problem 13 and the file Cellphone. In XLMiner's Partition with Oversampling procedure, partition the data so there is 50 percent successes (churners) in the training set and 40 percent of the validation data are taken away as test data. Construct a logistic regression model using Churn as the output variable and all the other variables as input variables. Perform an exhaustive-search best subset selection with the number of best subsets equal to 2. Generate lift charts for both the validation data and test data.
a. Why is partitioning with oversampling advised in this case?
b. From the generated set of logistic regression models, select one that you believe is a good fit. Express the model as a mathematical equation relating the output variable to the input variables. Do the relationships suggested by the model make sense? Try to explain them.
c. Using the default cutoff value of 0.5 for your logistic regression model, what is the overall error rate on the test data?
d. Examine the decile wise lift chart for your model on the test data. What is the first decile lift? Interpret this value.